DNA sequencing on a USB stick-sized device, $US900

By Grant Jacobs 18/02/2012

This has to be the bioinformatics and genomics news of the day.

Genome sequencing typically features expensive big boxes connected to powerful workstations.

Here’s something the size of USB stick that is set to sell for between $US500-900. How long until DNA sequencing is commonplace?


There’s so much pouring out on ‘da intertubes’, as the slang goes. Below I have gathered some of the more informative early sources, feedback from twitter then articles.

My brief comments are in square brackets.

The original press release is available on the Nanopore company website.*

I’ve gotten the tweets below** from following the Advances in Genome Biology and Technology meeting using the #AGBT hashtag.***

Update: Should have added – this is a one-use disposable device.

Twitter feedback

  • Chris Gunter @girlscientist: “Eric Green asks if MinION will work on Mac and PC — Brown says he is Mac user to applause from room. #AGBT” [Source – in fact the machine illustrated on the company website (above) is a Mac laptop.]
  • Chris Gunter @girlscientist: Brown: “we had a quick look at RNA” can be read directly with no need for cDNA, data look same as DNA. [Source – would be good to learn more about this, sounds a little tentative – ?]
  • David Schlesinger @djschlesinger: MiniION available second half of this year [Source]
  • The Assemblathon @assemblathon: I think one of the biggest wins from the @nanopore announcement is the promise of ‘transparent pricing’. I.e. all institutions pay the same [Source – May be partly an effect of move to a consumer model, as opposed to ‘exclusive research device’ – ?]
  • Nico R. @notSoJunkDNA: Amen ’@omespeak: Most commonly overheard snippet at #AGBT right now: ‘it is amazing, but we have to see it work in practice’’
  • Omespeak @omespeak: Wondering when we shall see real world data from Oxford Nanopore? #AGBT
  • Leonid Kruglyak @leonidkruglyak: Great problem to have: “how will 100kb+ reads change your research?” bit.ly/wKUHBT #AGBT
  • Keith Bradnam @kbradnam: Reactions to @nanopore news. Biologists: cheer. Bioinformaticians: groan. System admins: weep. #AGBT [I’m not so sure about the need for bioinformaticians to groan. More thoughts on this later, maybe – time permitting.]
  • Oxford Nanopore @nanopore: of course! RT @kbradnam: @nanopore release also reveals that MinION syncs with iTunes, acts as a 3G modem, and has a fresh lemon scent #AGBT [Well…]
  • [With my own tongue-in-cheek: No comments from BGI on this news yet!]

Articles with more details

From GenomeWebNews “Oxford Nanopore to Begin Selling Two Low-Cost DNA Strand Sequencing Instruments this Year” [More details are available for premium (paying) subscribers.]

Follow-up with details at Nature News


GenomeWebNews reports that “Shares of Sequencing Instrument Makers Fall as Oxford Nanopore Unveils Platforms” [Full-text to registered users only.]

The GenomesUnzipped blog has thoughts on ’Making sequencing simpler with nanopores’, their take on the new Oxford Nanopore machines.

@sujaik: Keith Robinson adds a dash of realism to the #nanopore #agbt gushiness 🙂 ’Oxford Nanopore doesn’t disappoint’ [Edited to include title of post linked to rather than link. That also has to be one of the more original twitter aliases I’ve seen to date.]


As a local aside, there was a local company (in Dunedin, New Zealand) developing a nanopore-based sequencing machine several years ago. I’ve no idea what happened to it.

* Curses on the designer who thinks a dark background is a good idea. Similarly the graphic for the device was presented at full 3000×2000 pixel, scaled rather than as a, say, 300×200 image linking to the full image.

** I’d re-work the formatting of the tweets, but the WordPress editor is showing it’s limitations again and I’m not willing to fight it. Sigh.

*** Excuse my not giving full sources for all tweets. You can, of course, (re)discover them yourselves by searching.

0 Responses to “DNA sequencing on a USB stick-sized device, $US900”

  • ArsTechnica has another follow-on piece (not a bad summary for a concise piece, it focuses on the larger system that I didn’t cover):


    It’s worth reading the articles to see the points that make this technology useful, it’s more than just the size (for the mobile version) or just the speed (for the clustered version).

  • Those looking for a skeptical take, albeit one with obvious conflicts of interest, could try at Forbes:


    (Without knowing more details I couldn’t say either way on this, but perhaps it’s worth noting that most of his criticisms are about manufacture, delivery & disclosure, which are also business model issues rather than solely science/technology issues – ?)

  • Hi Grant

    The Dunedin company you are thinking of is now called Izon – http://www.izon.com/ – although they were called Australo when they first started. They have shifted theirfocus from sequencing to nanoparticle analysis (such as virus detection).

  • Robert,

    Thanks for that, their name wasn’t coming to mind. I remember them as Australo; I think I still have one of the Australo brochures somewhere in the office. I see that their head office is now based in Christchurch.

  • I don’t know about you but I’m VERY excited about not having to worry about assemblies (much). Genome assembly has to be one of the dullest bioinformatic jobs ever – I’m sure it’s a challenge and I’m very glad people do it, but it’s definitely not for me. We just brought some new compute and had a major debate over how much RAM we needed for the machines – the words “genome assembly” and “high memory” kept raising their ugly heads. I’m pleased we went with high CPU and modest memory instead.

  • I’ve long thought long-run reads would come in time, and hence have tried to keep reminding myself not to get overly interested in assembly algorithms. (They are very interesting challenges if you’re interested in developing algorithms. Running someone‘s method week in and week out might be another matter!)

    The problem doesn’t entirely go away, of course, it just changes. (At least until someone figures out how to reliably feed an entire chromosome through a nanopore!)

    It’s early days for these things, with some people calling for real-world data on the devices. With the current error rates, I can’t help wondering if (some) people will aim for a mix of short accurate reads, longer and somewhat error-prone reads – ?

    For whatever it worth, other problems involving all-against-all comparisons can (potentially) benefit from large RAM, particularly if large RAM-based data structures are used to try make the compute times tractable.

  • Doug’s earlier comment is here:


    Rather than have a discussion on this split over several threads, I’d encourage anyone interested to follow the link above and comment there.

    Readers might note I haven’t a lot of free time at this time (as I’ve mentioned as much in some footnotes) – I’ll try find time to come back to this, somehow, but “over there” not here.

  • For those with subscription access to Nature Biotechnology, this short paper (letter) offers some more details:


    I haven’t read the full letter (too little free time…) but the abstract (below) indicates that one contribution to their approach is to ‘racket’ the DNA passing through the pore to slow it down so that the nucleotides can be better distinguished. (I recall reading elsewhere that the bases in the Nanopore device are not identified per individual base, but in groups of three bases, which might explain the higher error rates reported in this abstract.)


    An emerging DNA sequencing technique uses protein or solid-state pores to analyze individual strands as they are driven in single-file order past a nanoscale sensor. However, uncontrolled electrophoresis of DNA through these nanopores is too fast for accurate base reads. Here, we describe forward and reverse ratcheting of DNA templates through the α-hemolysin nanopore controlled by phi29 DNA polymerase without the need for active voltage control. DNA strands were ratcheted through the pore at median rates of 2.5–40 nucleotides per second and were examined at one nucleotide spatial precision in real time. Up to 500 molecules were processed at ~130 molecules per hour through one pore. The probability of a registry error (an insertion or deletion) at individual positions during one pass along the template strand ranged from 10% to 24.5% without optimization. This strategy facilitates multiple reads of individual strands and is transferable to other nanopore devices for implementation of DNA sequence analysis.